Why some people are better at spotting AI imagery

A Vanderbilt University study explains why some people are better at detecting faces created by artificial intelligence. Researchers tested 500 people across two studies using the AI Face Test, which measures individual differences in the ability to distinguish between real and AI-generated faces. The study found that traditional factors such as intelligence, experience with AI, or specialized face recognition skills did not predict who could reliably tell real from fake. Instead, a person's object recognition ability was found to be a predictor. This ability has been linked to other skills such as identifying lung nodules in chest x-rays and recognizing musical notation. The researchers are conducting follow-up studies to understand the cues used by people who are better at spotting AI-generated faces, with the goal of using this information to help train others.